Improving the prediction of chemotherapy dose-limiting toxicity in colon cancer patients using an AI-CT-based 3D body composition of the entire L1–L5 lumbar spine

Chemotherapy dose-limiting toxicities (DLT) pose a significant challenge in successful colon cancer treatment. Body composition analysis may enable tailored interventions thereby supporting the mitigation of chemotherapy toxic effects. This study aimed to evaluate and compare the effectiveness of using three-dimensional (3D) CT body composition measures from the entire lumbar spine levels (L1–L5) versus a single vertebral level (L3), the current gold standard, in predicting chemotherapy DLT in colon cancer patients.

Retrospective analysis of 184 non-metastatic colon cancer patients receiving adjuvant chemotherapy was performed. DLT was defined as any occurrence of dose reduction or discontinuation due to chemotherapy toxicity. Using artificial intelligence (AI) auto-segmentation, 3D body composition measurements were obtained from patients’ L1–L5 levels on CT imaging. The effectiveness of patients’ 3D L3 body composition measurement and incorporating data from the entire L1–L5 (including L3) region in predicting DLT was examined.

Of the 184 patients, 112 (60.9%) experienced DLT. Neuropathy was the most common toxicity (49/112, 43.8%) followed by diarrhea (35.7%) and nausea/vomiting (33%). Patients with DLT had lower muscle volume at all lumbar levels compared to those without. The machine learning model incorporating L1–L5 data and patient clinical data achieved high predictive performance (AUC = 0.75, accuracy = 0.75), outperforming the prediction using single L3 level (AUC = 0.65, accuracy = 0.65).

Evaluating a patient’s body composition allowed prediction of chemotherapy toxicities for colon cancer. Incorporating fully automated body composition analysis of CT slices from the entire lumbar region offers promising performance in early identification of high-risk individuals, with the ultimate aim of improving patient’s quality of life.